Online Learning and Detection with Part-Based, Circulant Structure
Online Learning and Detection with Part-Based, Circulant Structure
复制标题
DOI：
10.1109/icpr.2014.725
复制
发表时间：
2014-08
期刊：
影响因子：
--
通讯作者：
Osman Akin;K. Mikolajczyk
中科院分区：
文献类型：
--
作者：
Osman Akin;K. Mikolajczyk
循环结构核（circular Structure Kernel， CSK）作为一种简单而高效的跟踪方法近年来被引入。在本文中，我们提出了CSK的扩展，明确地解决了原始CSK遭受的部分遮挡问题。我们的扩展是基于零件的方案，提高了鲁棒性和定位精度。此外，我们通过将CSK纳入在线学习和检测框架来提高其长期跟踪的鲁棒性。我们对最近推出的八种跟踪方法进行了广泛的比较。我们的实验结果表明，当与在线学习方法相结合时，所提出的方法显着改善了原始的CSK，并提供了最先进的结果。
Circulant Structure Kernel (CSK) has recently been introduced as a simple and extremely efficient tracking method. In this paper, we propose an extension of CSK that explicitly addresses partial occlusion problems which the original CSK suffers from. Our extension is based on a part-based scheme, which improves the robustness and localisation accuracy. Furthermore, we improve the robustness of CSK for long-term tracking by incorporating it into an online learning and detection framework. We provide an extensive comparison to eight recently introduced tracking methods. Our experimental results show that the proposed approach significantly improves the original CSK and provides state-of-the-art results when combined with online learning approach.